Fabrication effects on Niobium oxidation and surface contamination in Niobium-metal bilayers using X-ray photoelectron spectroscopy
Tathagata Banerjee, Maciej W. Olszewski, Valla Fatemi · 2026
Read Fabrication effects on Niobium oxidation and surface contamination in Niobium-metal bilayers using X-ray photoelectron spectroscopy by Tathagata Banerjee, Maciej W. Olszewski, Valla Fatemi (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2601-21953.txt.
Introduction
Tathagata Banerjee, Maciej Olszewski, and Valla Fatemi examine the surfaces that limit superconducting devices. Resonators and qubits lose performance through dielectric losses in surface oxides. Labs fight those oxides with metal capping layers, surface passivation, and acid treatments.
The team studies how fabrication steps affect oxidation and contamination in bilayers of niobium with another metal. They use X-ray photoelectron spectroscopy, which reads the chemistry of the topmost atomic layers. Their 2026 preprint connects processing choices to the surface states that follow. Quieter surfaces mean longer-lived qubits, so this kind of careful materials bookkeeping is part of how quantum hardware slowly gets better.
Catalog record
| CREATOR | Tathagata Banerjee, Maciej W. Olszewski, Valla Fatemi |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 44 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2601-21953
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Fabrication effects on Niobium oxidation and surface contamination in Niobium-metal bilayers using X-ray photoelectron spectroscopy Tathagata Banerjee, Maciej W. Olszewski, Valla Fatemi Abstract Superconducting resonators and qubits are limited by dielectric losses from surface oxides. Surface oxides are mitigated through various strategies such as the addition of a metal capping layer, surface passivation, and acid processing. In this study, we demonstrate the use of X-ray photoelectron spectroscopy (XPS) as a rapid characterization tool to study the effectiveness cap layers for niobium for further device fabrication. We non-destructively evaluate 17 capping layers to characterize their ability to prevent oxygen diffusion, and the effects of standard fabrication processes – annealing, resist stripping, and acid cleaning. We downselect for resilient capping layers and test their microwave resonator performance. I Introduction Superconducting qubit devices have seen rapid advancement in performance metrics with integration of improved materials and fabrication procedures McRae et al. 2020 ; Murray 2021 . Recent results demonstrate that conventional transmons are a promising candidate for quantum computing Bal et al. 2024 ; Tuokkola et al. 2025 ; Bland et al. 2025 ; Chang et al. 2025 , with the surface oxide of the superconducting metals such as niobium (Nb), tantalum (Ta), and aluminum (Al) as the suspected source as a major source of loss Burnett et al. 2016 ; de Leon et al. 2021 ; Crowley et al. 2023 ; Murthy et al. 2025 ; Krasnikova et al. 2025 . Nb, in particular, is chemically resilient, has a high critical temperature, and grows in superconducting structure on many substrates; however it has a non-self-limiting lossy oxide Murthy et al. 2025 ; Murthy et al. 2022a ; Romanenko et al. 2020 ; Bafia et al. 2024 ; Wenskat et al. 2022 ; Altoé et al. 2022 . Different methods of mitigating the oxide have been reported, including encapsulation Bal et al. 2024 ; De Ory et al. 2025 ; Zhou et al. 2024 ; Karuppannan et al. 2024 , surface passivation Alghadeer et al. 2023 ; Zheng et al. 2022 , and acid treatments Verjauw et al. 2021 ; Gingras et al. 2025 . To this end, superconducting resonator losses serve as a rapid turnaround characterization technique for determining the performance potential of materials for qubit applications McRae et al. 2020 , while additional materials characterization techniques are required for understanding the origin of the loss. In this study, we characterize the effect of multiple standard processes conducted during fabrication – annealing, resist stripping, and acid cleaning – on superconducting bilayers composed of niobium with a cap layer. We disentangle the various effects by measuring the oxidation state of Nb with X-ray photoelectron spectroscopy (XPS), and leverage the depth sensitivity of the XPS, as shown in Figure 1 . While the X-rays penetrate several microns into the sample, the electrons escape from the top 5-10 \mathrm{nm} , as determined by the inelastic mean free path of electrons in various materials Powell 2020 . By having a thin encapsulation layer, we can non-destructively measure the Nb-metal interface and determine the amount of oxidized Nb. Figure 1: Schematic of X-rays entering the measured sample. Scattered electrons escape from 5-10 \mathrm{nm} of depth. XPS scan of 5\text{\,}\mathrm{nm} film will capture electrons from both the metal cap and Nb. We report on the efficacy of various metals, nitrides, and alloys as oxygen diffusion barriers, both as deposited and after annealing. We examine the surface contamination and damage after processing samples with an Integrated Micro Materials AZ 300T chemical resist strip bath Olszewski et al. 2025 . We also subject the Nb-metal bilayers to various acid cleaning processes, using Nanostrip, hydrofluoric acid (HF), and buffered oxide etch (BOE) Place et al. 2021 ; Torres‐Castanedo et al. 2024 . Finally, we fabricate resonators using a subset of the Nb-metal bilayers to examine the effect of the best caps on the resonator performance. II Materials and methods II.1 Sample preparation Samples were sputtered in an AJA International, Inc. magnetron sputtering chamber, as used in Olszewski et al. Olszewski et al. 2025 . We deposited 60\text{\,}\mathrm{nm} Nb films on intrinsic high-resistivity ( \geq 10,000 \Omega -cm) float zone silicon (100) 100\text{\,}\mathrm{mm} wafers followed by in-situ sputtering of the 5\text{\,}\mathrm{nm} capping layer. We analyze metals – Al, Au, Hf, Mo, Pd, Pt, Ta, Ti, W, and Zr; nitrides – NbN, ZrN, HfN, TiN, and TaN; and alloys – Ti-W and Zr-Y. Full details of the sample preparation is provided in Appendix A . The critical temperatures (T c ) of the samples were measured to ensure that the T c had not been reduced significantly from that of Nb, 9.24\text{\,}\mathrm{K} (Appendix B ). II.2 XPS sample processing Samples were subjected to three different processes: annealing , strip bath , or acid cleaning , conducted on the day of the XPS measurements. All XPS measurements were conducted in a ThermoScientific Nexsa G2 Surface Analysis system with a chamber pressure around 2 \times 10 -7 \mathrm{T}\mathrm{o}\mathrm{r}\mathrm{r} . Further system details are provided in Appendix C . The measurements were conducted on un-etched surfaces of the bilayer films to examine the amount of oxide in the Nb. We tested the robustness of the capping layers with two different annealing tests: air and vacuum. Annealing is a common practice in device fabrication, and high temperatures are used for hydrocarbon removal. We conducted both experiments at 200\text{\,}\mathrm{\SIUnitSymbolCelsius} for 1\text{\,}\mathrm{h} with air-annealing done on a hot plate in the fume hood, and vacuum-annealing done in a tube furnace with pressure below <5\times 10^{-6} \mathrm{T}\mathrm{o}\mathrm{r}\mathrm{r} . The proclivity of capping layers to allow Nb oxidation at 200\text{\,}\mathrm{\SIUnitSymbolCelsius} is likely indicative of trace oxidation taking place at lower temperatures. This may not be easily detectable at lower thermal exposures due to the resolution of the XPS, and yet still be a source of loss. For the strip bath test, samples were first spin coated with Microposit S1813 resist and baked at 90\text{\,}\mathrm{\SIUnitSymbolCelsius} for 1\text{\,}\mathrm{min} . Samples were then soaked in a 80-90 \mathrm{\SIUnitSymbolCelsius} heated bath of as-received AZ 300T stripper for one hour. The bath was followed by sonication in isopropanol and deionized (DI) water, and rinses in DI water and isopropanol. Samples were blow-dried with ultra-high purity nitrogen. For acid cleaning tests, the samples were dipped in three acids: hydrofluoric acid (HF), buffered oxide etch (BOE), and Nanostrip. For HF, a 2% HF dip for 1\text{\,}\mathrm{min} was followed by rinsing in DI water and blow-drying. The BOE dip was done in a 10:1 BOE solution for 5\text{\,}\mathrm{min} , followed by rinsing in DI water and blow-drying. The Nanostrip clean was done in a 30\text{\,}\mathrm{min} soak in 70\text{\,}\mathrm{\SIUnitSymbolCelsius} Nanostrip solution (a mixture of sulfuric acid, peroxymonosulfuric acid, and hydrogen peroxide), followed by rinsing in DI water and blow-drying. II.3 Resonator measurements The fabrication process of the resonators is detailed in Appendix G , and follows the fabrication in Olszewski et al. 2025 . Resonators were fabricated using a design from the open-source Boulder Cryogenic Quantum Testbed containing eight branches per sample with target frequency of 4-8 \mathrm{GHz} Kopas et al. 2022 . Samples were measured in a Bluefors small diameter cryostat with a base temperature of 500\text{\,}\mathrm{mK} . For each resonance, we vary the probe frequency and fit the